My Life as a MOOC Dropout

Lost in a class of thousands, a MOOC enrollee yearns for a sense of community and shared learning.

• By Audrey Watters
• 05/03/13

I'm hardly a traditional college student, but I find myself taking university-level classes more and more these days--thanks to the growing number of MOOCs on offer. Plenty of ink has been spilled about how these free online courses will disrupt higher education by providing access to the best professors and classes in the world to anyone with web access.

As so often happens with ed tech, there's a gulf between the hype and the reality. While these MOOCs can tout huge enrollment numbers--some 160,000 for the Stanford Artificial Intelligence course that jump-started this recent trend--they don't have high completion rates. Just 21 percent of enrollees actually completed that AI class, and only 14 percent passed.

People fail to complete these courses for lots of reasons, but it's important to highlight one in particular: What makes these courses so easy to sign up for makes them just as easy to drop. MOOCs have no barriers to entry--no fees, no prerequisites, no textbooks. If you bail out, there's no record on your transcript or financial aid to pay back. You come, you go.

I should know, because I'm beginning to feel like a serial MOOC dropout. I was one of the hundreds of thousands who signed up for the Stanford MOOCs when they launched. I registered for both Machine Learning and Artificial Intelligence. The former was taught by Andrew Ng, who has since launched Coursera. The latter was taught by Sebastian Thrun, who went on to start Udacity. But one look at the introductory assignments--Machine Learning's first lesson had the word "regression" in the title--told me that I was in way over my head. (It's been almost 20 years since I took Introduction to Statistics.) I never went back to the website for either class.

I was more hopeful about my chances in Udacity's Introduction to Computer Science (CS 101) course, which I figured would be much closer to my skill level. I've been learning how to program for about a year, but I've struggled with a number of online tools. It was high time for me to take a class instead.

I did fairly well for the first few weeks, dutifully watching the lecture videos and completing the homework assignments. I rather liked the design of the class: Each week's materials included 20 to 40 short videos (two to five minutes each), followed by short quizzes that didn't count toward the grade but checked to see if you understood the concepts.

As a computer science novice, I confess there were lots of times when I didn't grasp the concepts. Turning to the course's forums for peer support, I was frustrated to find that many of my fellow "students" were boasting that they were actually professional Python developers. (A recent survey of students in MITx's Circuits and Electronics 6.002x course found that 80 percent of respondents had taken a comparable course before. And a survey of students in Ng's Machine Learning course found that 41 percent already worked in the field.

The forums for my course weren't filled with warm-and-fuzzy "We're all in this together" camaraderie. "I can code that homework assignment in one line!" was more typical of the kind of posts that appeared. I should have responded, "Help! I can't figure it out," but the bravado in the discussion forums was off-putting. I stayed quiet. I dropped out.

The Motivation Factor
I was absolutely determined to complete my next MOOC, Coursera's version of CS 101. Frankly, it was only my bloody-minded determination that made the difference between my dropping out of Udacity's version and sticking with Coursera's equivalent. Both platforms utilize video lectures, although Udacity's tend to be shorter--Coursera's run as long as 20 minutes. They both set short quizzes and homework assignments that are graded by robo-graders. I found these automated grading systems incredibly frustrating, since they'd mark questions wrong that I knew were right (my coding may not have been elegant, but the program ran okay).

The discussion forums for both platforms were similarly overwhelming. I mostly steered clear of them while taking Coursera's CS 101. In a class of tens of thousands, I felt very much alone. Nonetheless, I completed the course--the only MOOC I've ever finished. It was really a matter of my motivation to finally see a MOOC through to its end.

What does this all mean for MOOCs as a delivery vehicle for education? Frankly, I'm not sure if that level of motivation is scalable. I don't mean just for me. I mean for all the students who sign up with hopes of success only to fall by the wayside. Do these large online classes engage and support and encourage learners? Or will only the most motivated thrive?

And what's at the heart of such motivation? Job skills? The hope for certification (if not accreditation)? In my case, none of the above. What keeps me most motivated--other than sheer will--is learning with and from others. I enjoy being part of a community in a classroom, and these new MOOC platforms are really struggling to create that sense of community.

Coursera's use of peer assessment is particularly problematic. An alternative to robo-graders, peer assessment turns over the evaluation of assignments to other students in the class. It was used in two classes I dropped (Internet History, Technology, and Security and Fantasy and Science Fiction), but the system was plagued with problems: Many students weren't native English speakers, many didn't understand what good essays entail, and there was no way to rate the feedback. With MOOCs, we've scaled the capacity for class size but we haven't scaled our social learning relationships.

To do so isn't impossible. Indeed, we need only hark back to the origins of MOOCs and the theory of connectivism from which they sprang. We can build out massive online learning networks; we can learn with and from each other.

I'm experiencing this now in another MOOC. The class, Ed Startup 101, is taught by David Wiley, associate professor of instructional psychology and technology at Brigham Young University (UT) and an open online course pioneer.

Ed Startup 101 has no video lectures, no multiple-choice quizzes, no robo-graders--and no closed LMS. Instead, students are asked for questions to help guide discussions with guest speakers. We blog on our own sites, where entries are then tagged and aggregated via RSS into the main course website and class blog. The emphasis isn't solely on the content. The emphasis is on the community of learners.